Join our community of smart investors

Further reading: Dealing with radical uncertainty

Lord Mervyn King and Professor John Kay teach us how to approach decision making
August 6, 2020

A global pandemic has always been a possibility. For years, governments around the world have had plans about what to do should a virus like coronavirus strike. Yet as the crisis unfolded, it quickly became clear that plans fell miles short of what was needed. The best economic model in the world could not have forecast when the pandemic would strike, and so when it came, we were not prepared.

The new book by Mervyn King, former Governor of the Bank of England, and John Kay, a leading British economist, could hardly have been more timely. Radical Uncertainty: Decision-making for an unknowable future, published in March, explains why relying on probabilistic reasoning via computer models to forecast the future is unsuitable and sometimes harmful. There are events that are 'radically uncertain' that we will never be able to predict.  

Rather than treating what comes out of models as if they were truths about the world, where invented numbers offer false security, the authors say we should construct robust narratives to understand ‘what is going on here’. Such narratives should yield confidence to manage uncertainty, and are based on our knowledge of context and our ability to interpret it.  

In an eerily foreboding manner, the book acknowledges in the third chapter, "we must expect to be hit by an epidemic of an infectious disease that does not yet exist. To describe catastrophic pandemics, or environmental disasters, or nuclear annihilation, or our subjection to robots, in terms of probabilities is to mislead ourselves and others.” On this basis, it can be argued that institutional over-reliance on models led us to underestimate the risk of a pandemic, and consequently left us ill prepared. 

Translating the critique of modelling to the financial world, Messrs King and Kay give a number of examples of how basing decisions on probabilistic forecasts alone can have dire consequences. Regulation, they argue, is an area where measuring risk through modelling has been particularly damaging. This was clear in the collapse of Northern Rock, which was deemed in spring 2007 to be the “best capitalised” bank in the UK under Basel regulations, a global regulatory framework for the banking sector.

By February 2008, the bank had collapsed. Regulators were wrong to believe that the risk facing the bank could be encapsulated in fixed numerical weights, loosely based on historical experience. “The possibility that funding from wholesale markets would simply become unavailable, and the investor appetite for mortgage-backed securities would abruptly diminish, was not considered,” the book points out. Northern Rock was felled by an off-model event.   

The authors also say the misuse of models has materially reduced the prospects of a secure retirement for many people. This followed a change in rules after the late Robert Maxwell looted the Daily Mirror’s pension fund in the 1990s. Pension regulations introduced in the UK in 2004 – post Maxwell – require all final-salary schemes to compute a technical valuation of liabilities over the life of the scheme – likely to be over 50 years. Thus, the contributions required to make sure the scheme is not in deficit are unaffordable to most companies – and consequently there are very few final-salary schemes left in the private sector. 

So, what’s the solution? Lord King and Professor Kay suggest robust principles to guide regulation, not tens of thousands of pages of detailed rules, which elevate the duty of compliance over the spirit of proper stewardship of people’s money. Readers might look on the collapse of Woodford Investment Management and sympathise with this approach. 

The authors don’t rebuke the use of models completely and conceded that many have proved indispensable. Harry Markovitz’s modern portfolio theory, for example developed in the 1950s taught us that risk should not be based solely on individual assets, but also on the relationship between assets – otherwise known as the benefits of diversification. 

But models should not be used to describe the world as it really is. Appreciation of what we cannot know, alongside what we learn from financial data, will make us better investors. The book invites readers to look at the achievements of Warren Buffett, George Soros and Jim Simons, each of whom has amassed tens of billions of dollars. They all have distinct styles but they share a common humility. 

“Buffett and Soros repeatedly emphasise the limits of their knowledge, and Simons claims he will never seek to override his algorithms,” the authors say.  “And all ignore – are even contemptuous of – the corpus of finance theory based on portfolio theory, the capital asset pricing model, and the efficient market hypothesis”. Indeed, said ‘corpus’ would imply they could not have succeeded as they have. 

The lesson of experience is that there is no single approach to financial markets that makes money or explains ‘what is going on here’, no single narrative of ‘the financial world as it really is.’ There is a lot that we cannot know. But the authors encourage us not to be frightened of uncertainty, as it is essential for innovation and true creativity.

Rich in history, maths, philosophy, psychology and often humour, there’s a lot of wisdom imparted from these two titans of economics in what looks like their valedictory tome.

 

Further Reading:

Radical Uncertainty: Decision-making for an unknowable future by Mervyn King and John Kay